feat: AI Property-Based Testing, Feedback System, Code Search, and Best Practice Recommendations

[KidDev88]

Closes #569, Closes #523, Closes #518, Closes #516

Type of Change

• New feature (non-breaking change that adds functionality)

---

Summary

This PR implements four AI-powered features for the StarForge CLI:

1. AI Property-Based Testing (AI Property-Based Testing #569): Automatically discovers contract properties and invariants, generates proptest-compatible test strategies, validates invariants across state transitions, discovers edge cases via boundary analysis, and provides shrinkage strategies for counterexample minimization.

2. AI Feedback and Learning System (AI Feedback and Learning System #523): Collects user feedback on AI-generated responses, tracks response quality metrics, learns user preferences from corrections, and generates improvement plans based on feedback trends.

3. AI Code Search and Discovery (AI Code Search and Discovery #518): Provides natural language code search, discovers patterns in the codebase, finds similar code blocks, and generates usage examples for contract functions.

4. AI Best Practice Recommendations (AI Best Practice Recommendations #516): Analyzes Soroban contracts against security, gas optimization, code organization, testing, deployment, error handling, storage, and access control best practices, with priority-ranked recommendations.

Motivation / Context

Closes #569, Closes #523, Closes #518, Closes #516

These features enhance the StarForge CLI with intelligent analysis capabilities:

• Property-based testing catches edge cases that unit tests miss
• Feedback system enables continuous improvement of AI responses
• Code search helps developers discover patterns and reuse code
• Best practice recommendations guide developers toward secure, optimized contracts

Changes

New Command Modules

• src/commands/ai_property_test.rs: CLI for property testing (discover, generate, validate, edge-cases, shrink)
• src/commands/ai_feedback.rs: CLI for feedback (record, stats, preferences, quality, improve, prompt)
• src/commands/ai_search.rs: CLI for code search (search, similar, patterns, usage, discover)
• src/commands/ai_recommend.rs: CLI for recommendations (analyze, scan, category, plan)

New Utility Modules

• src/utils/ai_property_testing.rs: Property discovery, invariant extraction, strategy generation, test case generation, shrinkage strategies
• src/utils/ai_feedback.rs: Feedback storage, preference learning, quality metrics, feature statistics
• src/utils/ai_search.rs: Code indexing, relevance scoring, pattern discovery, usage example generation
• src/utils/ai_recommendations.rs: Security, gas, testing, deployment, and access control analysis

Modified Files

• src/main.rs: Added command variants and handlers for all 4 features
• src/commands/mod.rs: Registered new command modules
• src/utils/mod.rs: Registered new utility modules
• src/commands/optimize.rs: Fixed unclosed delimiter bug

Testing

All new modules include unit tests:

• ai_property_testing: Property discovery, invariant extraction, strategy generation, test pipeline
• ai_feedback: Feedback recording, quality metrics, preference learning, prompt building
• ai_search: Code relevance scoring, function name extraction, search configuration
• ai_recommendations: Security checks, score calculation, prompt generation

Tradeoffs

• Local heuristic analysis provides fast results without Ollama, with optional AI enhancement
• Each feature follows existing CLI patterns (clap subcommands, structured output)
• No new dependencies required (uses existing proptest, serde, anyhow, colored)

Out of scope

• Real-time file watching for code search updates
• Cloud-based pattern sharing across projects
• Advanced ML models for pattern recognition (uses heuristic + optional Ollama)

[drips-wave]

@KidDev88 Great news! 🎉 Based on an automated assessment of this PR, the linked Wave issue(s) no longer count against your application limits.

You can now already apply to more issues while waiting for a review of this PR. Keep up the great work! 🚀

Learn more about application limits